The Class I Phosphoinositide 3-kinases (PI3Ks) regulate a wide range of cellular functions, including growth and survival, metabolism, lipid and protein synthesis, and motility. Of the Class I enzymes, only PI3K? is regulated both by receptor tyrosine kinases (RTKs) and G-protein coupled receptors (GPCRs), the latter through direct binding of the PI3K? catalytic subunit (p110?) to G?? subunits. We recently provided the first mechanistic information about GPCR regulation of PI3K?, by identifying and mutating the p110? binding site for G??. Surprisingly, many of the cellular activities of PI3K?, including its abilityto transform fibroblasts and to support the growth of PTEN-null tumor cells, require its interactions with G?? subunits. Moreover, when we replace endogenous p110? with a GPCR-uncoupled mutant in breast tumor cells, the cells show decreased in vitro 3D invasion, decreased tumor growth in an orthotopic model, and decreased metastasis in an in vivo metastasis assay. Notably, disrupting GPCR inputs to PI3K? caused a greater decrease in tumor growth than eliminating PI3K? lipid kinase activity. These data suggest that GCPR coupling to PI3K? plays critical roles in tumor development and metastasis. This coupling could provide an important new drug target for cancer chemotherapy. In addition to G??, p110? also binds the small GTPase Rab5. We identified and mutated the Rab5 binding site in p110?. Cells expressing Rab5-uncoupled p110? show pronounced defects in macropinocytosis, a clathrin-independent endocytic process used by tumor cells to obtain nutrients that support rapid growth. These data suggest that Rab5-PI3K? interactions could provide a novel drug target for some tumors. Finally, published studies have shown that the PTEN tumor suppressor forms a complex with p110? in cells, whereas in vitro data shows that PTEN preferentially binds dimers of the p85 regulatory subunit. These data appear to be incompatible, as p85 is not thought to dimerize when bound to p110 catalytic subunits. We have developed novel tools for manipulating the multimeric state of p85, which we can use to define how p85 and p110? interact with PTEN. Specific Aim 1 examines GPCR signaling to PI3K? in mouse models of prostate and endometrial cancer, using conditional and whole animal knock-ins of mutant p110?. Aim 2 proposes mechanistic studies on the regulation of macropinocytosis by Rab5-PI3K? interactions, and examines the role of Rab5-p110? binding in dendritic cell function and in a K-Ras-driven model of pancreatic cancer. Finally, Aim 3 uses analytical ultracentrifugation, which measures the molecular weight of oligomers independently of shape, to define interactions between PTEN and PI3Ks. Taken together, we have identified novel mutants, made novel biological tools, and devised new experimental strategies, to better define the role of PI3K? in cell biology and human disease.